A New Multiple Classifier System for Diagnosis of Erythemato-Squamous Diseases Based on Rough Set Feature Selection

The effectiveness of classification and recognition systems has improved in a great deal to help medical experts in diagnosing diseases as a second opinion. In this paper, we develop a diagnosis model based on a combination of classifiers, known as a multiple classifier system, to diagnose erythemato-squamous diseases. The proposed model consists of two major stages. First, rough set-based feature selection is used to select the optimal feature subset from the original feature set in order to both improve the accuracy and shorten the response time of our classification. Second, an ensemble of three classifiers including MLP, KNN and SVM is created to make their own decision on selected features independently; eventually, majority voting method is used to combine the obtained results from each classifier and return the final decision of this intelligent system as a diagnosis result. Experimental results show that the proposed ensemble model achieves 97.78% classification accuracy using 12 selected features of the erythemato-squamous diseases dataset taken from UCI (University of California at Irvine) machine learning database. Therefore, we can conclude that our proposed method is very promising compared to the previously reported results.